←back to thread

Ask HN: How are you productively using Claude code?

23 points nocobot | 1 comments | 13 Jul 25 08:28 UTC | HN request time: 0.216s | source

I tend to be sceptical when it comes to LLM based coding tools but many people seem to be raving about huge productivity gains which I wouldn’t mind as well.

However when trying cc it left me vey disappointed. For context I’m working on a relatively greenfield rust project and gave it tasks that I would consider appropriate for a junior level colleague like:

- change the return type of a trait and all it’s impls

- refactor duplicate code into a helper function

- replace some of our code with an external crate

it didn’t get any of them correct and took a very long time. Am I using the tool wrong?

How are you using cc or other agentic tools?

1. sometimes_all ◴[14 Jul 25 18:12 UTC] No.44563375[source]▶
The more open-source code the LLM has read (ie is in its training data), the better it will be. Rust is not a very popular language compared to others, so Claude might not be able to perform as well as it does with, say, Python or Javascript. Also, the more recent models might be better than the older ones.

I've been having fun with Claude Code and VSCode's agent. Any reasonably experienced engineer should be able to use it for a subset of languages without too many issues, but they definitely need to hydrate the context (eg. using Claude.md) and have a sensible set of system prompts set up. Good, well-written and broken-down-into-steps user prompts are non-negotiable.